ABSTRACT Antimicrobial resistance is now recognized by world leaders, including the US Congress (and at least one recent President) as one of the most serious public health threats to mankind. Our proposal focuses on Enterococcus faecium (Efm), an extremely resistant, hospital-associated (HA) Gram-positive coccus that now accounts for almost 40% of nosocomial enterococcal infections, some of which are medically untreatable. >80% of clinical Efm isolates are vancomycin resistant (VRE) and almost all of these are ampicillin resistant (Amp-R), the traditional drugs of choice; resistance to newer agents (linezolid, daptomycin) is well documented and increasingly reported. We now know that most clinical Efm isolates belong to a distinct Efm clade (the HA clade A1) that is thousands of years separated from the community-associated (CA) clade B (comprised mostly of human commensals). A medically very important phenotypic difference is that CA human commensals (clade B) are readily inhibited by Amp (MICs 0.125-2 mg/L), while HA isolates are Amp-R (MICs often >64 mg/L). PBP5 of Efm is a low-affinity penicillin- binding protein (PBP) that is essential for Amp R; it comes in two main forms referred to as PBP5-S (CA clade B, Amp-S strains) and PBP5-R (HA clade A1, Amp-R strains); PBP5-S/R appears to be a transitional form found in clade A2 (animal clade) that is closely related to clade A1. pbp5-R and pbp5-S alleles differ by ~ 5% and there is a consensus of 20-21 amino acids (aa) (some PBP5-Rs have an extra aa) that can distinguish -S from -R proteins; using a multi-copy plasmid, engineered changes of 4 aa, most near the predicted active site, significantly lowered affinity of the corresponding recombinant PBP5 for [14C]penicillin and increased Amp MICs. However, studies of naturally occurring alleles or of mutated genes in the native chromosomal pbp5 site have not been reported. PBPs are well known targets for antibiotics and are generally amenable to structural studies. Our long-term overarching goal is to use PBP5 structure:function relationships to develop therapeutics against Efm, including VRE. Here, we propose to provide the background for future drug design that would target Amp-R Efm. 1). We will use X- ray crystallography to determine the high-resolution structures of various PBP5-S, PBP5-R and intermediate PBP5- S/R proteins that impact function, such as ?-lactam MICs (after cloning pbp5 alleles into the native pbp5 site of a ?pbp5 HA Efm), and affinity of recPBP5 proteins encoded by the alleles for ?-lactams. A comparison of high- resolution structures of PBP5 forms plus co-crystallization with ?-lactams +/- peptidoglycan components, functional assays and specific site-directed pbp5 mutations, should provide the foundation for understanding the role of specific aa changes seen in nature on affinity of PBP5 forms for ?-lactams and provide insights into how to inhibit this PBP's function. Future projects (e.g., an R33) would then use these structures to rationally design or improve the efficacy of inhibitors of PBP5 forms that confer Amp-R to develop an effective anti-Efm therapy. 2). We will test our hypothesis that Psr of clade B Efm, but not the truncated Psr found in clade A1, results in repression of PBP5 expression, preventing Amp-R even in the presence of pbp5-R, with a possible future goal of designing Psr mimickers.